Search papers, labs, and topics across Lattice.
GreenSeg, a novel ground segmentation algorithm, was developed to enable autonomous navigation for agricultural robots in challenging Mediterranean greenhouses using low-cost RGB-D sensors. It combines robust global plane fitting with a surface curvature filter for terrain adaptability and a seed-point-based region growing constraint to ensure spatial continuity. Experimental results on the AGRICOBIOT I platform demonstrate that GreenSeg outperforms benchmark methods, achieving up to 19.24% improvement in mIoU, particularly during rotational maneuvers.
A low-cost RGB-D perception framework can now enable reliable autonomous navigation for agricultural robots in complex greenhouse environments, outperforming existing methods by up to 19% in mIoU.
Greenhouse agriculture in the Mediterranean region faces significant automation challenges due to its unique structural and environmental constraints. These environments are characterized by extremely narrow aisles, heterogeneous terrains ranging from concrete to tilled soil and severe optical interference caused by polyethylene covers, which induce specular reflections and "ghost points" in depth sensors. While autonomous navigation is essential for digitizing agricultural tasks, traditional solutions often rely on expensive 3D LiDAR systems that are economically unscalable for most facilities. To address this, this paper presents GreenSeg, a robust perception framework for autonomous navigation using RGB-D sensing. The proposed method introduces a dual-layer validation strategy: a robust global plane fitting combined with a surface curvature filter for terrain adaptability, and a seed-point-based Region Growing constraint to ensure the spatial continuity of the navigable plane. Experimental validation was conducted using the AGRICOBIOT I platform across four diurnal scenarios with varying solar elevations. The results show that GreenSeg consistently outperforms benchmark segmentation methods, achieving peak improvements of 11.58% in mean Recall and 19.24% in mIoU during critical rotational maneuvers at the end of corridors. These findings confirm that the proposed algorithm enables stable and safe autonomous navigation in unstructured, dynamic agricultural environments that are subject to budget constraints and sensitive to lighting conditions.
